Building a Banking Data Pipeline: From Raw Data to KPIs #16

In Part 16 of the Banking Data Pipeline series, we turn our processed banking customer data into business KPIs using SQL aggregations in Amazon Athena: total customers, customers by country and customers by segment. Each KPI is one short query on the customers table in our S3 data lake, built with COUNT(*), GROUP BY and ORDER BY. We also look at why a row count and a count of distinct customer IDs aren't always the same KPI, and why the business and data teams should agree the definition. All customer data shown is synthetic test data. Chapters 0:00 Intro 0:49 Opening Athena on the processed customers table 1:45 KPI 1: total customers with COUNT(*) 3:21 Row count vs distinct customer IDs 3:32 KPI 2: customers by country with GROUP BY 5:32 KPI 3: customers by segment with ORDER BY 7:56 Your turn: segment percentages 8:39 What's next Next up — Part 17: Validating and Optimizing Athena Queries. Previous — Part 15: Querying Athena Tables on S3 Data with SQL: [ADD LINK] Full playlist: [ADD LINK] Which SQL KPI would you build next? Tell me in the comments. #AWS #AmazonAthena #DataEngineering